Contaminated Confessions: How Source and Consistency of Confession Details Influence Memory and Attributions

Via "contamination," false confessions usually contain accurate and nonpublic details, and details that are inconsistent with the case facts (Garrett, 2010). In two studies (N1 = 476; N2 = 364), we replicated previous findings that inconsistent confessions yield fewer guilty verdicts than accurate confessions (Henderson & Levett, 2016; Palmer et al., 2016). The source of the details in a confession (interrogator vs. suspect) also influenced guilt decisions: when details were introduced by the suspect, participants were more confident in his guilt than when the interrogator introduced the details. Exploratory analyses revealed that although consistency and source of confession details affected memory for crime facts, memory did not influence verdict decisions as much as participants' causal attributions for the confession. Consistency and source influenced whether participants attributed the confession to the suspect's guilt or something other than guilt (e.g., coercion), which in turn affected guilt judgments and confidence. General Audience Summary Sometimes innocent people confess to crimes they did not commit. Often, false confessions contain accurate, nonpublic crime details that only the police and the true perpetrator could have known. How is it possible for an innocent person to know intimate details of the crime? Police interrogators may, purposely or inadvertently, communicate details about the crime during the interrogation-details that the suspect can then repeat in the false confession. This process is known as contamination. Importantly, false confessions also usually contain mistaken details that do not match the evidence collected by the police. In two studies, we examine whether people notice (1) when interrogators or suspects introduce important crime details during a confession and (2) when the details in the confession match or do not match the actual crime facts. We found that participants do notice. When a confession included accurate details that came from the suspect, participants were most likely to say that the suspect was guilty. However, this is not necessarily because they accurately remember what the details were or where they came from. Instead, people appear suspicious of bad confessions because they seem involuntary and they believe that the suspect confessed not because of his guilt, but because the interrogator forced him to confess. Research on contamination is important because the contaminated details can make a confess